Isoform specific Interactome Analysis of Spastin

Abstract

Hereditary Spastic Paraplegias (HSPs) are a heterogeneous group of inherited neurodegenerative disorders, that are distinguished by an axonopathy of the upper motor neurons and therefore clinically present with a spasticity and weakness of the lower limbs. Complicated forms of the disease can include additional symptoms such as cognitive impairment, ataxia or myopathy (Klebe et al., 2015). HSPs can be inherited in an autosomal recessive, dominant or X-chromosomal manner. The most common form of autosomal dominant HSP is caused by a mutation in the Spastic Paraplegia Gene 4 (SPG4), encoding for the protein spastin, which was first described in 1999 (Bürger et al., 2000, Solowska and Baas, 2015). Four isoforms of the protein, that are shown to differ in their cellular expression levels and localization (Claudiani et al., 2005, Solowska and Baas, 2015), are known to be endogenously expressed. Furthermore, spastin integrates several key pathways of HSP pathogenesis, including membrane shaping, cytoskeleton dynamics as well as intracellular transport and is known to interact with a number of HSP-associated proteins such as Atlastin or REEP1 (Evans et al., 2006, Park et al., 2010). While there have been many postulations about the disease mechanism of SPG4, such as a loss of function of the protein (Solowska et al., 2010) or the toxicity of the truncated M1 spastin isoform (Solowska et al., 2017), it was shown that not all disease-causing mutations can be explained by these theories. Therefore, up to today, the pathomechanism remains uncertain. In this work, a mass spectrometry-based approach was chosen to perform an isoform-specific interactome analysis of spastin. The Flp-InTM T-RexTM system was used to create stable SH-SY5Y overexpression cell lines for the four endogenously expressed spastin isoforms. The tagged protein was then isolated by immunoprecipitation and bound interaction partners were identified by mass spectrometry. Promising interaction candidates were subsequently confirmed in co- immunoprecipitation studies. Abstract 76 Resulting from this workflow, we were able to reveal the two novel protein-protein interaction partners of the SPG4 protein spastin, NUP43 and ATP5A. Our findings indicate an interaction of the longer M1 isoform of spastin with the NUP107-160 complex, a subunit of the nuclear pore complex, that is known to play a major role in the assembly of the nuclear pore complex and is presumed to promote the spindle assembly during mitosis. Surprisingly, another finding was the interaction of spastin with proteins of the F1 subunit of the mitochondrial ATP synthase, such as ATP5A. As an impairment of mitochondrial functions was previously shown for other forms of HSP (e.g. SPG7,13), an affection seems possible. As those novel M1 spastin interactions were identified in a simplified cell model after cell lysis, they will need to be confirmed a second in vivo cell model, for example through co-localization studies. Furthermore, the relevance of these possible spastin interactions in post-mitotic neuronal cells requires further investigation. A better understanding of the spastin function in health and disease will hopefully bring us closer to revealing the disease mechanism in SPG4 and the development of treatment options. Abstract